Bug 8517 - 2.6.21 regression - system hang under heavy load with "lapic"
Summary: 2.6.21 regression - system hang under heavy load with "lapic"
Status: REJECTED INVALID
Alias: None
Product: ACPI
Classification: Unclassified
Component: Power-Thermal (show other bugs)
Hardware: i386 Linux
: P2 normal
Assignee: Len Brown
URL:
Keywords:
: 8518 (view as bug list)
Depends on:
Blocks:
 
Reported: 2007-05-21 03:45 UTC by Gregory Krzystek
Modified: 2007-08-06 18:48 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.21
Subsystem:
Regression: ---
Bisected commit-id:


Attachments
dmesg (16.35 KB, text/plain)
2007-05-24 01:12 UTC, Gregory Krzystek
Details
acpidump (85.57 KB, application/octet-stream)
2007-05-24 01:12 UTC, Gregory Krzystek
Details

Description Gregory Krzystek 2007-05-21 03:45:15 UTC
Most recent kernel where this bug did *NOT* occur: 2.6.20
Distribution: Gentoo 
Problem Description:
when thermal zone is enabled in kernel and module is loaded when system is
overloaded (glibc compilation or other long compilation process) system hangs
Steps to reproduce:
compile glibc on thremal zone enabled kernel
Comment 1 Gregory Krzystek 2007-05-21 03:50:40 UTC
*** Bug 8518 has been marked as a duplicate of this bug. ***
Comment 2 Len Brown 2007-05-23 18:46:56 UTC
What model machine is this?

Did this work properly in all releasese tested before 2.6.21?

Please attach the output from acpidump
Please attach the complete output from dmesg -s64000

Please paste the contents of /proc/acpi/thermal_zone/*/* here

Please monitor the /proc/acpi/thermal_zone/*/temperature
and
/proc/acpi/fan/*/*
and report what you see before the system hangs.
I assume it is rising and the hang occurs at some trip-point
or perhaps a thermal failure.

Is the fan spinning at full speed when the failure occurs?
Comment 3 Gregory Krzystek 2007-05-24 01:11:10 UTC
>What model machine is this?
Laptop MAXDATA ECO 4200X

>Did this work properly in all releasese tested before 2.6.21?
it was work properly with kernels <= 2.6.20

>Please attach the output from acpidump
>Please attach the complete output from dmesg -s64000
>Please paste the contents of /proc/acpi/thermal_zone/*/* here
cooling mode:   active
<polling disabled>
state:                   ok
temperature:             41 C
critical (S5):           100 C
passive:                 92 C: tc1=2 tc2=3 tsp=100 devices=0xedec357c
active[0]:               68 C: devices=0xedece978
active[1]:               61 C: devices=0xedecea68
>Please monitor the /proc/acpi/thermal_zone/*/temperature
>and
>/proc/acpi/fan/*/*
>and report what you see before the system hangs.
>I assume it is rising and the hang occurs at some trip-point
no in kernel 2.6.20 when i compiling glibc cpu temp is 59-72C
on 2.6.21 system hang when temp is ~52-58C
>or perhaps a thermal failure.
>Is the fan spinning at full speed when the failure occurs?
full speed of fan i have when cpu temp is ~70C
on 2.6.21 system hangs before cpu temp is that high.
i will atth all info you need
sorry for my english
Comment 4 Gregory Krzystek 2007-05-24 01:12:00 UTC
Created attachment 11582 [details]
dmesg

full dmesg from 2.6.21
Comment 5 Gregory Krzystek 2007-05-24 01:12:42 UTC
Created attachment 11583 [details]
acpidump

acpidump output
Comment 6 Knut Petersen 2007-08-04 00:52:48 UTC
Try to _enable_ polling:

echo 2 > /proc/acpi/thermal_zone/THRM/polling_frequency

If that does not help, please attach disassembled dsdt.
Comment 7 Len Brown 2007-08-06 18:48:35 UTC
> Kernel command line: root=/dev/hda6 lapic ...

> Local APIC disabled by BIOS -- reenabling.
> Found and enabled local APIC!

Overriding the BIOS in this way is not (and can not be) supported.
Please re-open this bug report if you can reproduce the hang
without "lapic" on the command line.

Note You need to log in before you can comment on or make changes to this bug.